Related
I was just wondering if anyone had come across the Smart Wi-Fi app from Kineto that T-Mobile is purportedly testing now. Is there a more concrete timetable for release known. Also, I have heard conflicting information about the software requirements. I have read somewhere that it would run on any Android device with v. 1.5 or greater, but I have also heard that it requires special APIs that it needs to hook into so does that mean that the carrier would have to issue new builds of Android for each device with the APIs built in or did Google bake the APIs into Eclair or Froyo?
Anyone? I'm only gonna bump this once and if no one answers I'll take that as a confirmation that no one knows.
Ill throw down a bump on this.
Looks like they have officially put out the info. Now it is just waiting on T-mobile to push it out to users.
Funny thing is they say it is for Froyo, but Eclair is being showed on Vibrant as a demo.
http://www.kineto.com/press_releases/Kineto_Smart_Wifi_Android2_Froyo.php
Im kinda curious what this means exactly.
The press release doesn't seem to imply that this will be for one network or country, nor did the article that I read here.
Hopefully this will turn into something that'll be for any droid device. As someone who lives in a blackspot I would LOVE this kinda service. Here's hoping it gets into a release candidate soon. Wish they had opened it up to everyone rather than keeping in closed development tho, but oh well.
This would be a seriously amazing service and really boost android's rockingness.
I'd love to see if you could hack this to get true phone capability to wifi only tablets somehow *ponders*
I hope we get 2.2
http://it.slashdot.org/story/10/11/05/0229205/Researcher-To-Release-Web-Based-Android-Attack
"The attack targets the browser in older, Android 2.1-and-earlier versions of the phones."
http://forums.t-mobile.com/t5/Samsung-Vibrant/Security-vulnerability-in-2-1/td-p/535335
And the thread appears to have already been locked.
EDIT: My bad, the link icon isn't a lock icon.
What an ass. So he figures out something and now hes going to release it?
So is his intensions to piss people off or force Googles hands to fix it?
kizer said:
What an ass. So he figures out something and now hes going to release it?
So is his intensions to piss people off or force Googles hands to fix it?
Click to expand...
Click to collapse
I think its the latter. That, or to light a fire under the OEMs & network operators to get 2.2 out to more devices. Just my $0.02...
Sent from my SGH-T959 using XDA App
The current OEM vendor/carrier model is one of the worst parts of Android. Google attempted to break this model via the Nexus One. Hopefully it does light a fire to improve the security model for these phones.
Google may be forced to rein in some of the rampant variances to secure the platform via enforcing a minimum level of compliance to security updates or else revoke a phone makers ability to use the Android trademark.
The problem has already been fixed with 2.2, so the onus is on the OEMs to get their act together.
Some things make me want to respect this guy, then again it affects me since we have yet to recieve 2.2. But yes I believe all android phones should be running current software.
I wonder if you need to be rooted in order to fall the vicitm, unless you can push superuser.apk via the exploit and run it.
Have to give him props for trying, and like seeing that he is using linux based OS to develop on
lqaddict said:
I wonder if you need to be rooted in order to fall the vicitm, unless you can push superuser.apk via the exploit and run it.
Have to give him props for trying, and like seeing that he is using linux based OS to develop on
Click to expand...
Click to collapse
Youre right! Maybe he works for T-mobile and is secretely making all our phones go back to stock and unrootable. Which in turns means they will never release 2.2 hahaha. <- By the way do not take this as actual fact I know how the paranaoid are here on the forums lol
lqaddict said:
I wonder if you need to be rooted in order to fall the vicitm, unless you can push superuser.apk via the exploit and run it.
Have to give him props for trying, and like seeing that he is using linux based OS to develop on
Click to expand...
Click to collapse
No, this a generic exploit within WebKit. The actual exploit itself doesn't have superuser access, it can only access what the web browser is able to access. It can't make phone calls or generate SMS messages, but it can access files like photos and whatever else is available to non-rooted apps.
I don't know why you guys think this guy is a douche. This is how it always worked. When people find security vulnerbilities, they tell the company, but the company usually doesn't move it up to the top of the list to fix. So they mention the type of security flaw there is, sends the information to the company, and sometimes even mention it at conferences. After publicly announcing it, they give the company time to fix it, otherwise it's the company's fault for not getting their ass in gear to fix the security issue.
DKYang said:
I don't know why you guys think this guy is a douche. This is how it always worked. When people find security vulnerbilities, they tell the company, but the company usually doesn't move it up to the top of the list to fix. So they mention the type of security flaw there is, sends the information to the company, and sometimes even mention it at conferences. After publicly announcing it, they give the company time to fix it, otherwise it's the company's fault for not getting their ass in gear to fix the security issue.
Click to expand...
Click to collapse
I do no see how he is a douche.
Ignoring the issue does not make it disappear, and he clearly has done his work to make the issue public in hopes it gets addressed.
Releasing a code with a security hole that you have to use something to circumvent the security of the device to fix is douche (apple vs jailbreakme.com anyone)
kizer said:
What an ass. So he figures out something and now hes going to release it?
So is his intensions to piss people off or force Googles hands to fix it?
Click to expand...
Click to collapse
I was paranoid by this too. My Vibrant will shackled from having sex with the web until it gets 2.2 Maybe that researcher wants them to release Froyo soon so use this to leverage against them to release ASAP?
I don't think he's a douche. I honestly want to believe that google would push carriers to be on the same OS. Just the fact that not all android phones can handle the 2.2 OS - And so people stuck with those phones and would be affected by this flaw is pretty crappy. But I really hope this makes carriers want their phones updated and running the latest and greatest. Only time will tell.
I am thinking of getting one since it just went on sell at dell for $299. I would like not to mess with it and keep it at stock so wondering?
1. Gps with this phone? Works? Compared to google maps?
2. Can you allow pop ups with the browser?
3. Camera is ok?
Thanks for any help
everythiing you are asking pertains to WP7 in general with the exception of the camera; It sucks BIG TIME.
this is where megapixel count is stupid. the 5MP on DVP is just as bad as my motorolla V3.
Thanks for the info on the camera...i was looking everywhere and cant find info on if i can have pop ups for my work web site. Anyone know? Any info on gps nav would be great
I don't think popups are disallowed. This, however, is not the same as being allowed. Confusingl, huh?
As far as GPS, there are several turn-by-turn apps in the marketplace ranging from free to $6. Plus, in Mango there will be onboard navigation as part of the OS. We'll see if it remains tap-by-tap or if it goes to a turn-byturn system.
Otherwise, great phone especially with the newer firmware.
That is confusing for the pop ups lol......i really need them for work so hopefully i can get it to work. Is mango an update? When will it come out?
Thanks for your help
Yes. Mango is an update due out in September/October. A tweet hit yesterday which may have leaked a September release. But, we'll see. Meanwhile, here's a massive review of Mango for you to pour over.
http://pocketnow.com/windows-phone/the-ultimate-windows-phone-7-mango-preview
It might be sooner..
http://www.wpcentral.com/toshiba-fujitsu-mango-phone-slated-august-release-japan
If that is true (i have my doubts) i'd find it hard to believe that japan would get mango before the US.
Hi everyone, I'm not sure how useful this thread will be - but I hope it will be useful. As I've watched new N10 users come into the fold of finding and wanting to learn more about "the reboot problem" I realize the information on this topic is spread all over the place. I've tried to gather the information, as well as present it in a good fashion, such that there is a link/thread that can be easily referenced. I'm not the most technical of this crowd (I mean c'mon I am here at XDA!) so please forgive me and notify me of any needed corrections or clarifications.
---[Documenting] Google's Nexus 10 MALI GPU memory leak *UNRESOLVED*---
The Nexus 10, henceforth N10, is an Android tablet made by Samsung for Google. At launch in November 2012 the N10 was shipped with Android 4.2.0
*
Included in the N10 is a SOC (system-on-a-chip) from the Exynos line:
http://en.wikipedia.org/wiki/Exynos_(system_on_chip)
*
The Exynos uses a GPU made by ARM. The GPU component is the ARM Mali-T604.
http://en.wikipedia.org/wiki/Mali_(GPU)
*
Starting immediately after the N10’s launch there were reports of “random reboots”. At first not much was known about the exact cause of these reboots. The audience participating in these reports experienced varying degrees of testing results, especially as 4.2.1 and 4.2.2 updates rolled out. The majority of this discussion is captured in two threads:
https://productforums.google.com/forum/#!topic/mobile/VFYnt7uN9d0[1-25-false] *
and
http://forum.xda-developers.com/showthread.php?t=1998496
(*Note that the Google thread was erroneously marked ‘Answered’ after one of the OTA roll outs.)
*
The community was able to flush out that the memory/RSS use of Android’s surfaceflinger process seemed to be the victim of a memory leak, and that by monitoring the RSS value one could predict when the tablet would experience the degraded performance. Viktor.M then created this open issue report that focuses in on that:
https://code.google.com/p/android/issues/detail?id=54757
*
About that time a user named apstrand, who appears to either be an ARM developer or at the very least someone who is extremely knowledgeable in this area had found more exact and detailed info on this, and he posted it here:
https://code.google.com/p/android/issues/detail?id=52579
*
Matthew Fry began contacting Samsung, then in turn ARM, as he posted in one of the Google threads. Eventually he was able to get this response: "I can confirm that we were able to reproduce the issue in the stock Nexus 10 4.2.2 provided by Google. We were, however, unable to reproduce it with our release driver.
We will test the application again when the next release of the firmware for Nexus 10 is published."
*
From this point, knowing ARM has code that avoids the error, people began looking into working on the code themselves, and asking ARM for assistance. ARM has posted source code components for the T-6xx series of drivers here:
http://malideveloper.arm.com/develo...n-source-mali-t6xx-gpu-kernel-device-drivers/
However ARM has also stated, “these components are not a complete driver stack. To build a functional OpenGL ES you need access to the full source code of the Mali GPU DDK, which is provided under the standard ARM commercial licence to all Mali GPU customers.” This means Samsung has an ARM DDK (Driver Development Kit), but to date has not released it/aspects of it to the public.
*
For now there appears nothing anyone can do without access to the ARM DDK.
*
EDIT/UPDATE: Since my original post here, a lot of activity happened in a short period of time. Many comments below capture this. I will add the history into this thread, but the big news is 4.3 is out for the N10 now and the community is testing. Like each time a new Android version has rolled out, community testing is "all over the place" with different people reporting different results. People who know how to measure the exact parameters on this bug are reporting it looks like it may be fixed! Its noteworthy though that there is some suspicion of other issues. Because I want to keep this first post based only on facts, I am going to wait a while and see the community results before posting an update that cant be backed up by actual data.
***
FINAL UPDATE: Ok, looks like its fixed now. I'm not saying they didn't morph it somehow to another issue, but I am saying the exact situation we were all focused on here is no longer going on w/ the 4.3 (4.3.x if you want to include the 3mb update.)
bigmatty said:
Hi everyone, I'm not sure how useful this thread will be - but I hope it will be useful. As I've watched new N10 users come into the fold of finding and wanting to learn more about "the reboot problem" I realize the information on this topic is spread all over the place. I've tried to gather the information, as well as present it in a good fashion, such that there is a link/thread that can be easily referenced.
Click to expand...
Click to collapse
Nice overview of the problem. I would add that one of the Android bugs you link to is marked as assigned, by none other than Jean-Baptiste Queru, the technical lead of the Android Open Source Project at Google. He also renamed the other issue, so it appears that Google is finally aware of the problem and working on it. Hopefully, this means a fix is forthcoming in the upcoming Android 4.3 release, but that is merely a hope on my part: I have no inside info.
bigmatty said:
Hi everyone, I'm not sure how useful this thread will be - but I hope it will be useful. As I've watched new N10 users come into the fold of finding and wanting to learn more about "the reboot problem" I realize the information on this topic is spread all over the place. I've tried to gather the information, as well as present it in a good fashion, such that there is a link/thread that can be easily referenced. I'm not the most technical of this crowd (I mean c'mon I am here at XDA!) so please forgive me and notify me of any needed corrections or clarifications.
Click to expand...
Click to collapse
Thank you, this thread was sorely needed. This seems like a really big mess-up on Google's part. Has there been a memory leak this bad before for any other device?
patsmike said:
Thank you, this thread was sorely needed. This seems like a really big mess-up on Google's part. Has there been a memory leak this bad before for any other device?
Click to expand...
Click to collapse
I don't think so. My Dad has a Nexus 7 and it doesn't have this issue at all. I really hope that 4.2.3 or 4.3 fixes this - it's making me completely lose patience with the device and I'm getting tempted to sell it. First the light leak issue and now this. Come on Google, you really are taking the piss here. I'm not sure how they could get it so wrong?!
bigmatty said:
...This means Samsung has an ARM DDK (Driver Development Kit), but to date has not released it/aspects of it to the public.
Click to expand...
Click to collapse
So this fix requires Samsung to do it.
Almost hate to say it but I really don't see it being fixed anytime soon...
I confirm that surfaceflinger is f***ed up on my Nexus 10 too.
It screws up very, very fast when connected with USB debugging and I start to code OpenGL ES 2.0-based live wallpapers.
---------- Post added at 10:04 PM ---------- Previous post was at 09:58 PM ----------
espionage724 said:
So this fix requires Samsung to do it.
Almost hate to say it but I really don't see it being fixed anytime soon...
Click to expand...
Click to collapse
Not that bad. If it was HTC, we'd be screwed forever. With Samsung, we have a tiny (well, like, really tiny ) hope that something will be done in this regard.
I have been reading that Nexus owners may get 4.3 this month - anyone think the problem may be addressed?
In this thread http://productforums.google.com/forum/#!topic/mobile/VFYnt7uN9d0[476-500-false] - some dude from Google acknowledges that they know it's a problem - but he hasn't said anything in 6 months.
Really hope it is sorted out - I am keeping my fingers crossed.
bruce_aisher said:
I have been reading that Nexus owners may get 4.3 this month - anyone think the problem may be addressed?
In this thread http://productforums.google.com/forum/#!topic/mobile/VFYnt7uN9d0[476-500-false] - some dude from Google acknowledges that they know it's a problem - but he hasn't said anything in 6 months.
Really hope it is sorted out - I am keeping my fingers crossed.
Click to expand...
Click to collapse
As a comment above suggests, the bug has been assigned by Jean Baptiste-Queru little more than a week ago. We can only hope that it was fixed and merged into 4.3.
starkilling said:
As a comment above suggests, the bug has been assigned by Jean Baptiste-Queru little more than a week ago. We can only hope that it was fixed and merged into 4.3.
Click to expand...
Click to collapse
From what I read the developer of the surface finger driver already has an in-house fix and acknowledged that they reproduced the issue with the stock ROM. Those two comments give me hope.
Sorry but I do not have a link. I also have no idea why fixed drivers have not been released considering every ROM (stock and custom) must use the same proprietary drivers. It is likely that they require Google to distribute the fix.
I really hope to see this fixed soon, the random reboots really make the tablet unreliable. I can't do much serious work on the thing since every couple of hours or so the thing resets, losing all my work.
tbosje said:
I really hope to see this fixed soon, the random reboots really make the tablet unreliable. I can't do much serious work on the thing since every couple of hours or so the thing resets, losing all my work.
Click to expand...
Click to collapse
I use RasBeanjelly and I have not had a single random reboot. No reboots on mrRobinson's AOKP as well. You do have to reboot every couple of days, though. It really depends from ROM to ROM. Stock really rebooted on me like a motherf***er though.
I saw that one of the Google staff responded to the google product support thread the other day, a sign of progress?
starkilling said:
I use RasBeanjelly and I have not had a single random reboot. No reboots on mrRobinson's AOKP as well. You do have to reboot every couple of days, though. It really depends from ROM to ROM. Stock really rebooted on me like a motherf***er though.
Click to expand...
Click to collapse
I've spent so much time customising my apps and launcher, flashing my whole unit is such a hassle. Guess I might have to though :/
tbosje said:
I saw that one of the Google staff responded to the google product support thread the other day, a sign of progress?
I've spent so much time customising my apps and launcher, flashing my whole unit is such a hassle. Guess I might have to though :/
Click to expand...
Click to collapse
You can back them up, you know.
Posted by Matthew Fry on Google Product forums earlier today :good:
Ladies and gentlemen, I have the most amazing news. Pretty much the best news I can think of.
I have finally received word back from the Mali Developers. They revisited the issue and found that the fault was indeed in the Mali drivers.
"Following previous emails, the Nexus 10 issue that you have been mentioning has been fixed in the latest versions of the Mali driver."
It is, unfortunately, Google's responsibility to release this new driver version with the next Nexus 10 factory image. We can only hope that the fact that I/O has come and gone and the recent release of the Google Edition S4 and One on 4.2.2 does not mean that JB 4.3 or KLP are a long ways out.
chinmaysk said:
Posted by Matthew Fry on Google Product forums earlier today :good:
Ladies and gentlemen, I have the most amazing news. Pretty much the best news I can think of.
I have finally received word back from the Mali Developers. They revisited the issue and found that the fault was indeed in the Mali drivers.
"Following previous emails, the Nexus 10 issue that you have been mentioning has been fixed in the latest versions of the Mali driver."
It is, unfortunately, Google's responsibility to release this new driver version with the next Nexus 10 factory image. We can only hope that the fact that I/O has come and gone and the recent release of the Google Edition S4 and One on 4.2.2 does not mean that JB 4.3 or KLP are a long ways out.
Click to expand...
Click to collapse
Link, please? I can't find it.
Edit: found it, nvm. there were lots of pages
Can one of the mods please pin this thread in the forum? It seems like every week someone comes in and asks about this issue. Once a fix is released, this thread can be unpinned.
6 Month and Samsung and Google havent resolved that.....
Hope 4.3 comes out soon
kswiss86 said:
6 Month and Samsung and Google havent resolved that.....
Hope 4.3 comes out soon
Click to expand...
Click to collapse
Well, I hope the driver didn't come out from Mali too late to make the 4.3 release.
Funny thing is I even said to myself that I'll never buy anything from Samsung again, how wrong I was thinking the Nexus 10 would be different because it's from Google.
With 4.3 out next week (possibly) I do hope Google decide to put this emergency fix in. It really gets on my nerves that this tablet just randomly reboots because I know how much potential this tablet has and I feel this is holding it back.
Sent from my Nexus 10 using xda app-developers app
Just install steel series controller drivers or steel series engine!
http://steelseries.com/g/steelseries-engine
Will totally brick your windows installation! Good times, thank you for the loss..
I've tested this 3 times on a new installation, Had to reimage and reinstall, no luck. can't use game pad controller driver, won't upgrade..
Congrats, you're a beta tester. Have you actually reported this to anybody whose job it is to do something about it?
It isnt called an early build for no reason. Its out here for you to find those issues and report them to microsoft, not ***** on forums about it.
Whoa, beta software being unstable?! Somebody call Microsoft!
Wow, beta software being put on the windows store without being call beta software!, it's a pre-release, NOT BETA.
But, it's good that your so eager to point that out.
So, have you reported the bug yet?
Also, exactly what do you believe the distinction is between "pre-release" and "beta"? I mean, http://windows.microsoft.com/en-us/windows-8/preview-faq sure sounds like "beta" to me even if they don't use that actual word.
Testing with older or obscure software and hardware is the primary reason why Microsoft (and other large software companies) have pre-release versions, whatever they're called. You're in a position to help both Microsoft and yourself, here.
With MS, public pre-releases are for PR, for aficionados to get the software and do the evangelizing gruntwork. It's not a beta, because there is no formal mechanism for bug reporting, despite the "send it in" button. That, and normal people don't know how to properly beta test anyway. A pre-release is not supposed to have any obvious bugs, although peripheral driver compatibility is among the last things to be tested.
These said, 8.1 is the first of MS' new "rapid release" product, so common sense would indicate it's gonna be rougher than the old 3-yr cycle. From the smattering of reports, some features are still unfinished.
With RTM 1-2 months out, it's probably not worth the trouble to install the pre-release, unless you have a spare box to throw it onto, or you're a 0-day guy who GOTTA HAVE IT NOW.